GEO Accession viewer

NCBI > GEO > Accession Display

Not logged in | Login

GEO help: Mouse over screen elements for information.

Sample GSM2779089

Query DataSets for GSM2779089

Status

Public on Oct 01, 2018

Title

HL-E115-DelHoxc-Mm-cHiC-2

Sample type

SRA

Source name

Embryonic Limb Bud

Organism

Mus musculus

Characteristics

strain: C57BL/6
tissue: Hindlimb
developmental stage: E11.5
genotype: DelHoxc

Growth protocol

Forelimb and hindlilb buds were micro-dissected from E11.5 mouse embryos

Extracted molecule

genomic DNA

Extraction protocol

3C-libraries were prepared from homozygous E11.5 forelimb and hindlimb buds as described previously (Hagege et al. 2007). cHi-C experiments were performed as quadruplets (WT) or duplicates (mutants). Per biological replicate, 5-6 pairs of limb buds (ca. 3 ×106 cells) were micro dissected in PBS at room temperature (RT). A single cell suspension was obtained by incubating the tissue in 500 µl Trypsin-EDTA 0.05 % (Gibco) at 37°C for 10 minutes shaking at 900 RPM. The cells were resuspended and homogenised using a 0.40 µm cell strainer (Falcon) and diluted in 10% FCS/PBS. Cells were fixed by adding 650 µl 37% formaldehyde (Sigma-Aldrich) with a final concentration of 2% and mixed for 10 minutes at room temperature. Fixation was quenched using 1.425 M glycine (Merck) on ice and immediately centrifuged at 260 g for 8 minutes. Supernatant was removed, the pellet resuspended in lysis buffer (final concentration of 10 mM Tris pH 7.5, 10 mM NaCl, 5 mM MgCl2, 0.1 M EGTA, and 1x Complete protease inhibitors (Roche)) and incubated on ice for 10 minutes. Removal of lysis buffer was done by centrifugation at 400 g for 5 minutes at 4°C, followed by removal of supernatant, snap-freezing and storage at -80°C. On the next day, the pellet was resuspended in 520 µl 1x DpnII buffer (Thermo Fisher Scientific), and incubated with 7.4 µl 20% SDS shaking at 900 RPM at 37°C for 1 hour. Next, 75 µl 20% x-100 Triton was added and left shaking at 900 RPM at 37°C for 1 hour. A 15 µl aliquot was taken as a control for undigested chromatin (stored at -20°C). The chromatin was digested using 40 µl 10 U/µl DpnII (Thermo Fisher Scientific) shaking at 900 RPM at 37°C for 6 hours. 40 µl of DpnII was added and samples were incubated overnight shaking at 900 RPM at 37°C. On the third day, 20 µl DpnII was added to the samples shaking 5 more hours at 900 RPM at 37°C. DpnII restriction enzyme was inactivated at 65°C for 25 minutes. A 50 µl aliquot was taken to test digestion efficiency (stored at -20°C). Next, the digested chromatin was diluted and re-ligated in 5.1 ml H2O, 700 µl 10x ligation buffer (Fermentas), 5 µl 30 U/µl T4-ligase (Fermentas), incubated at 16°C for 4 hours and shake manually 3 times. The ligated samples were incubated further 30 minutes at RT. The chimeric chromatin products and test aliquots were de-crosslinked o.n. by adding 30 µl and 5 µl Proteinase K, respectively, and incubated at 65°C overnight. On the fourth day, 30 µl or 5 µl of 10 mg/ml RNase was added to the samples and aliquots, respectively, and incubated for 45 minutes at 37°C. Next, chromatin was precipitated by adding 1 volume phenol-chloroform to the samples and aliquots, vigorously shaking them, followed by centrifugation at 4000 RPM at RT for 15 minutes. The upper phase containing the chromatin was transferred to a new tube. To the aliquots, 100% ethanol was added and the samples were frozen for 30 minutes, centrifuged at 5000 RPM for 45 minutes at 4°C, washed with 70% ethanol and resuspended in 20 µl 10 mM Tris-HCl pH 7.5. To the samples, 7 ml H2O, 1 ml 3M NaAc pH 5.6 and 35 ml 100% ethanol was added. The samples were frozen at -20°C for 3 hours. The precipitated chromatin was isolated by centrifugation at 5000 RPM for 45 minutes at 4°C. The chromatin pellet was washed with 70% ethanol, and further centrifuged at 5000 RPM for 15 minutes at 4°C. Finally, the 3C-library chromatin pellet was dried at RT and resuspended in 10 mM Tris-HCl pH 7.5. To check the 3C-library, 600ng was loaded on a 1% gel together with the undigested and digested aliquots.
The 3C-library was then sheared using a Covaris sonicator (duty cycle: 10%, intensity: 5, cycles per burst: 200, time: 6 cycles of 60 s each, set mode: frequency sweeping, temperature: 4 to 7 °C). Adaptors were added to the sheared DNA and amplified according to Agilent instructions for Illumina sequencing. The library was hybridized to the custom-designed sure-select beads and indexed for sequencing (50 to 100 bp paired-end) following Agilent instructions.

Library strategy

OTHER

Library source

genomic

Library selection

other

Instrument model

Illumina HiSeq 4000

Data processing

Library strategy: HiC
Preprocessing and mapping of paired-end sequencing data, as well as filtering of mapped di-tags was performed with the HiCUP pipeline v.0.5.8 (Wingett et al. 2015) (Nofill: 1, no size selection, Format: Sanger). The pipeline used Bowtie2 v.2.2.6 (Langmead and Salzberg 2012) for mapping short reads to reference genome (NCBI37/mm9). Replicates were combined after mapping and filtering. Filtered di-tags were further processed with Juicebox command line tools to bin di-tags (5 and 10 kb bins) and to normalize the map by Knight-Ruiz (KR) matrix balancing (Lieberman-Aiden et al. 2009; Knight and Ruiz 2013; Durand et al. 2016)+ (Durand et al, Cell Systems 3(1), 2016.). For this, only reads with a MAPQ ≥30 were considered. The DNA-capturing step enriches genomic region chr13:53,400,001-57,300,000 on mm9 leading to three different regimes in the cHi-C map: (i) enriched versus enriched, (ii) enriched versus non-enriched, and (iii) non-enriched versus non- enriched. For binning and normalization only di-tags in regime (i) were considered. Therefore di-tags were filtered for the enriched region and mm9 coordinates were shifted by 53,400,000 bp. For Juicebox, a custom chromosome sizes file containing only the enriched region on chr13 (length 3900000 bp) was used. After binning and normalization, coordinates were shifted back to their original values. All maps were processed on the WT reference genome to work with the same genomic coordinates across all samples. To account for differences between maps in their distance dependent signal decay, maps were scaled pairwise by the sum of their sub-diagonals. Therefore, each sub-diagonal vector in one matrix is divided by its sum and multiplied by the average of the sums of both matrices. To normalize for sequencing depth, each map was additionally converted to reads per million (RPM). To avoid copy number biases, a region spanning all tested deletions (chr13:55,730,001-56,500,000) was not considered for the computation of scaling factors, for the diagonal normalization as well as for the RPM normalization. cHi-C maps of count values and subtraction maps were visualized as heat maps truncating all values above the 99-th percentile for visualization purposes. The percentile was determined from absolute values within the same region used for normalisation.
Differential interactions were determined from the pairwise subtraction of normalized maps. For diagonal and RPM normalization, as well as for the computation of P-values, individual regions spanning the corresponding deletions were excluded. To avoid artefacts from KR normalization due to low coverage, maps were analysed on 10 kb resolution and a single row with very low coverage (chr1:54,450,001-54,460,000) was excluded. To account for the distance dependence of the magnitude of differences, each difference value was subtracted by the mean and divided by the standard deviation of the corresponding sub-diagonal. For the computation of mean and variance per sub-diagonal, values above the 99.5-th percentile were not included. P-values for the z-transformed difference values were computed using a standard normal distribution and further corrected for multiple testing with false discovery rate (Benjamini and Hochberg 1995). In histograms the distributions of z-scores do not fit the normal distribution perfectly, but show as well a bell-like shape.
To obtain more fine-grained interaction profiles, we generated virtual Capture-C-like profiles based on the same filtered BAM files also used for the cHi-C maps and defined several virtual viewpoints of 10 kb size. A read pair was considered in the profile, when one mate mapped to the defined viewpoint region and the other one outside of it. Reads were counted per restriction fragment and binned further to a regular 1kb grid. In case a fragment spanned more than one bin, the count value was distributed proportionally to the overlaps. Afterwards, the profiles were smoothed by averaging over a sliding window of 5 bins. For coverage normalization the profiles were divided by sum of counts on chr13 and multiplied with 103. The region ± 5kb around the viewpoint as well as the regions in Tab. XX_DEL were mutually excluded from the computation of the scaling factor in pairwise comparisons. The profiles were generated with custom Java code using htsjdk v1.139 (https://samtools.github.io/htsjdk/).

Submission date

Sep 11, 2017

Last update date

May 15, 2019

Contact name

Guillaume Andrey

E-mail(s)

guillaume.andrey@unige.ch

Phone

+41223795703

Organization name

University of Geneva